Paper to be presented at the 35th DRUID Celebration Conference 2013, Barcelona, Spain, June 17-19 Institutional voids as a trigger for the emergence of born global production and innovation networks Tarmo Kalvet Tallinn University of Technology Ragnar Nurkse School of Innovation and Governance [email protected] Helena Barnard University of Pretoria Gordon Institute of Business Science [email protected] Marek Tiits Institute of Baltic Studies [email protected] Abstract The article complements the existing work on the evolution of global innovation networks (GINs) from a less developed innovation context by documenting that GINs can also be ?born global?, rather than be the outcome of the gradual upgrading of global production networks into GINs. The article, focusing on two GINs (Skype and MXIT) that operate digital services, suggests the importance of institutional theory as a useful lens for understanding firms? actions. Both companies have emerged from countries with less developed innovation systems, Estonia and South Africa, and have developed GINs in response to the talent, product market and capital market voids of those countries. While the existing literature places a strong emphasis on firm capabilities, our evidence suggests that firms can internationalise their production and innovation also due to a lack of local capabilities. Jelcodes:O32,- Institutional voids as a trigger for the emergence of born global production and innovation networks Abstract The article complements the existing work on the evolution of global innovation networks (GINs) from a less developed innovation context by documenting that GINs can also be “born global”, rather than be the outcome of the gradual upgrading of global production networks into GINs. The article, focusing on two GINs (Skype and MXIT) that operate digital services, suggests the importance of institutional theory as a useful lens for understanding firms’ actions. Both companies have emerged from countries with less developed innovation systems, Estonia and South Africa, and have developed GINs in response to the talent, product market and capital market voids of those countries. While the existing literature places a strong emphasis on firm capabilities, our evidence suggests that firms can internationalise their production and innovation also due to a lack of local capabilities. Keywords: institutional void; global innovation network; Skype; MXIT; capacities; alliances. 1. Introduction This article documents the emergence of two global innovation networks (GINs) from a less developed innovation context. It complements the existing work on the evolution of GINs by documenting that GINs can be “born global”, rather than be the outcome of an evolutionary process, and suggests the importance of institutional theory as a useful lens for understanding firms’ actions. A GIN can be defined as a globally organized web of complex interactions between firms and non-firm organizations engaged in knowledge production related to and resulting in innovation. Ernst (2002, 2010) has pioneered work on global production networks (GPNs) and GINs, and his work, focusing on multinational corporations (MNCs) in East Asia, suggests that there is an evolution 1 from GPNs to GINs. We contribute an understanding of another process through which GINs emerge: We document GINs from two countries with less developed innovation systems, Estonia and South Africa, that are “born global” (Knight & Cavusgil, 2004) not only in terms of their markets, but specifically in terms of their production and indeed innovation. We find that the global production dimension and global innovation dimension emerge at the time that firms are founded, and simultaneously rather than in sequence. Both home countries of the firms we study have a number of institutions to meet the economic, technological and social needs of society, but as is typical of less developed contexts, those institutions are in many ways inadequate. The “institutional voids” – which is the situation in most emerging markets when specialist institutions and intermediaries are either completely absent or not functioning as well as they might (Khanna, Palepu & Sinha, 2005; Khanna & Palepu, 2010) – create both challenges and opportunities for firms. In this study of two digital service firms, Skype (from Estonia) and MXIT (from South Africa), we find evidence that firms from such countries respond to the institutional voids by developing GINs from the outset. 2. Literature review This literature review is divided into two sections. First we discuss what is known about the emergence of GINs, linking our discussion not only to the globalisation of innovation, but also to the literature on “born globals”. Second, we discuss institutional voids, and how that body of literature relates to locational attractiveness. 2.1 The globalisation of innovation In the international business literature, the highlighting of strategic asset seeking as a specific motive for internationalisation (Dunning, 1998) gave rise to an extensive body of work on how MNCs use their global spread to seek not only natural resources, efficiency or markets abroad, but also new sources of knowledge and innovation. Various strands of literature have been documenting the emergence of global innovation networks. Studies include work on the 2 internationalisation of R&D (Cantwell & Kosmopolou, 2002; Pearce, 1999) and more broadly innovation (Archibugi & Iammarino, 1999) and the literature on global production networks (Ernst & Kim, 2002; Henderson et al, 2002). The about two decades of intellectual engagement with the phenomenon has had different foci, but evolutionary theory has been the most important lens underlying these studies, allowing us to integrate the key themes from the existing body of work into a predominant view (see Figure 1). [Figure 1] One characteristic of the current literature is a fairly clear distinction between GPNs and GINs. This clear distinction is to some extent a function of the conceptualisation of innovation in much of the work on the internationalisation of R&D. A clear distinction between GPNs and GINs is useful to the extent that the Doing, Using and Interacting (DUI) mode of innovation that typically takes place during the production process is considered to be quite different from the Science, Technology and Innovation (STI) mode innovation that can emerge out of a formal R&D process (Jensen et al, 2007). The bias towards studying the STI mode of innovation is also likely to lead to a greater emphasis on MNCs from the technologically and economically most advanced countries, since they are the largest private R&D actors globally. Because MNCs with their global reach can relatively easily orchestrate global networks, and are therefore more likely to be central nodes of the GINs than smaller or standalone firms, for long the bulk of research on the internationalisation of innovation emphasised the intra-firm networks of MNCs, i.e. their subsidiaries (Cantwell & Mudambi, 2005; Driffield, Love & Menghinello, 2009; Gupta & Govindarajan, 2000; Zander, 2002). As the field has evolved, there has increasingly been an interest in the many actors outside the MNC with which MNCs interact to innovate, and work increasingly also documents the contribution of non-firm partners of MNCs (e.g. Andersson, Forsgren & Holm, 2002; Chesbrough, 2006; De Jong, Kalvet & Vanhaverbeke, 2010; Frost, Birkinshaw & Ensign, 2002; Nell, Ambos & Schlegelmilch, 2011; Van de Vrande, de Jong, Vanhaverbeke & Rochemont, 2009). 3 Given that the networks that enable innovation are not necessarily intra-firm, but can also consist of various extra-firm and even non-firm partners, a GIN can in principle be initiated by standalone firms as much as by MNCs. The common equation of GINs with MNCs (e.g. Sachwald, 2008) is therefore not inherent in the definition of a GIN, but instead a result of the fact that relatively few stand-alone firms have the capabilities to coordinate production and innovation through a non-firm network. Moreover, it seems that when non-MNCs engage in GPNs/GINs, they follow a different process than MNCs do. 2.2 Born-globals and the development of business networks The work on “born-globals” provides a useful lens to understand “born global production / innovation networks”. In their pioneering 2004 article, Knight and Cavusgill define born-globals as firms (typically with less than 300 employees) that generate more than 25% of their turnover from exports less than three years after inception. A number of studies highlight the use of networks as an important mechanism used by born-globals to access both markets and capabilities (Presutti, Boari & Fratocchi, 2007; Sharma & Blomstermo, 2003). The emphasis of both the original Knight and Cavusgill paper and subsequent studies (e.g. Fan & Phan, 2007; Hashai, 2011; Lu, Zhou, Bruton & Li, 2010) has been on the globalization of markets, with the size and scope of the market abroad as the key indicator in defining whether or not a firm should be considered born global. However, we propose that a firm can also be born global in terms of its production and innovation functions. Such a firm is from the outset organised into a global network of partner organisations that help the firm to develop, distribute and innovate its offering. Such born global innovation networks often have highly globalised markets, but not necessarily. Instead, what characterises these firms is that they from the outset develop their offering through extra-firm networks.